The result of a study of 1130 aftershocks of the 1980 Campania-Lucania earthquake are presented. The data were collected primarily with a dense array of portable smoked-drum seismographs deployed in the epicentral region a few days after the main shock. Tests performed on subsets of the data to establish the location accuracy are described in detail, and epicentral maps and cross-sections are presented and discussed. One hundred and sixty well-constrained fault plane solutions using upward traveling rays are presented. The distribution of aftershocks forms an elongate cloud which does not define a plane. This is not because the events are poorly located. A subset of events located with epicentral errors of less than 1 km and depth errors less than 2 km forms a similar cloud to the complete aftershock set. The aftershock mechanisms are very varied. Some are similar to the main event but many differ profoundly. These include events on shallow or vertical planes and thrust events. The latter include some events which require contraction in the direction in which the main event indicates extension. The aftershocks of all mechanisms are intermixed in space in such a way that no simple system of faults can explain them. A model to explain the aftershocks and the lack of surface ruptures in the main event is offered. It is proposed that the main shock cuts between two aseismic detachment surfaces and that the aftershocks are a consequence of the complicated deformation that such a geometry demands.
Whatever interpretation is adopted, the results are a challenge to the commonly accepted view that aftershocks even approximately define simple fault planes or that aftershock mechanisms replicate the mechanism of the main shock.
Defects in interpretations of the main shock and aftershocks process arise largely from a lack of information about the geometry of structures at depth in the epicentral region. Suitable deep reflection profiles could help to resolve this problem.